Asthma and reactive airway disease are the most common chronic medical conditions in children and affect over 7 million in the US alone. Asthma prevalence tends to be highest in inner cities and has a disproportionate impact on minorities. Adverse respiratory effects in children with asthma are associated with exposures to nitrogen dioxide (NO2) - an air pollutant with both indoor and outdoor sources. Principal indoor sources include gas stoves as well as other gas appliances, natural gas-fueled home heating, and infiltration from outdoor sources (e.g., traffic). Over one-third of US households use gas for cooking, and in homes with sources, highest indoor NO2 concentrations are found in inner-city residences where gas service is common and house volumes are smaller compared to suburban homes. Data from epidemiologic studies indicate that a reduction in NO2 concentrations commonly encountered in urban homes would result in a clinically significant reduction in asthma severity. We propose a randomized, double-blind, triple cross-over intervention study to examine the efficacy of substantial reduction of indoor NO2 on reducing asthma severity. To carry out an optimal Phase III clinical trial (randomized, double-blind, cross-over), we have designed an air cleaning device to accommodate each one of three filter types: (1) an NO2 scrubber composed of a filter that removes 80% of NO2 and also 80% of particles; (2) a nanofilter that removes all suspended particulate matter > 0.12 m (but removes less than 10% of the NO2); and (3) a control that filters neither NO2 nor particles. We propose to enroll 600 asthmatic children between 5-11 years of age living in homes with gas stoves and high NO2 levels and to randomly allocate them to a sequence of three experimental home interventions over a 16-week period while following them to assess asthma severity. Pilot testing suggests that two air cleaning devices operating at flow rates on the order of 450 cubic feet per minute, each with the NO2 scrubber in place, will ensure reductions in NO2 in a modest-sized home on the order of 90% or greater. The primary study outcome is a 5-level asthma severity score (Global Initiative for Asthma guidelines), calculated using symptoms, asthma medication use (recorded on calendars) and spirometry-assessed lung function (using NIOSH-certified, trained research assistants). The research aims are: (1) Determine whether reducing exposure to NO2 and particles compared to a control (non-filtered) condition results in a clinically significant reduction in asthma severity by one or more levels; and (2) Determine whether reducing exposure to particles > 0.12 m alone (without concurrent reduction in NO2) similarly reduces asthma severity. We focus recruitment efforts on asthmatic children from urban families most likely to have high indoor levels of NO2 (e.g., families with gas stoves living in multifamily housing). If trial results indicate that children with asthma would benefit from a reduction in exposure to indoor NO2, practical and economical remediation strategies are readily available.